资本资产定价模型及其扩展模型的实证比较研究

发布时间：2018-03-13 06:19

本文选题：资本资产定价模型　切入点：三因素模型　出处：《贵州财经大学》2013年硕士论文　论文类型：学位论文

【摘要】：资本资产定价模型（CAPM）作为现代金融财务理论的核心内容之一，首先由夏普（1964）、林特纳（1965）和莫辛（1966）在马克维茨的均值—方差理论的基础上分别提出，随后Black，Jensen和Scholes提出著名的BJS双程回归检验法、Fama-Macbeth的二阶段回归法、Roll的Roll批评和Fama-French提出的著名三因素模型,众多学者不断的对标准CAPM模型苛刻假设的放宽，逐步发展和形成了跨期资本资产定价模型(ICAPM)、基于下侧风险资本资产定价模型(D-CAPM)、三因素模型（Three-factor model）和回报贝塔法资本资产定价模型（Reward βCAPM）等拓展模型。 20世纪90年代初，中国上海证券交易所和深圳证券交易所先后成立，成立至今数20载已被运用到各行各领域中去。新兴市场相比国外资本市场，存在许多体制上的不完善，无法满足理论的假设条件，因此，CAPM模型在我国是适用性也有待考究。本文选取2003年-2012年上证180指数样本股的月收益率，将研究区间分为子区间T1从2003-2007年和子区间T2从2008年-2012年，在三因素分组和β值分组下运用时间序列回归和二阶段回归法对标准CAPM模型、下侧风险CAPM模型、FF三因素模型、回报贝塔法CAPM模型和基于下侧风险FF三因素模型进行综合比较分析以及对各组合各模型进行深入的风险收益关系的研究，得出以下结论：第一．计量系统风险β值的方法会影响模型对收益率的解释力和模型总体拟合度。我们知道β系数是衡量单个资产相对于市场的波动性的指标，对于β值的计算方法，国内外也有众多学者有研究，共同的结论是，即不同的方法会影响模型的整体效果。第二．在系统风险外，非系统风险对收益的影响是显著的。这可以从FF三因素模型和标准CAPM模型、下侧风险CAPM模型和回报贝塔CAPM模型的比较中看出，并且收益率的与规模因素成反比，与市值成正相关。第三．所有模型在我国股市的适应性都不强。在多次分组别，多方法的比较中，，标准CAPM模型和回报贝塔法CAPM模型的解释力基本相当，都不是很强，而下侧风险CAPM模型和FF三因素模型以及我们引入的基于下侧风险FF三因素模型虽然有所提高，但是整体的拟合度和解释力都不是很强。第四．除了对全样本收益风险研究外的大部分结论证明我国股市资金的时间价值为负值。说明我国资本市场还不够成熟、完善，相比国外成熟市场，国内股市的现状是信息披露不够充分，财务造假情况较多，重融资轻分红，信息不对称，噪声交易，受政策影响较大等的因素存在，加上投资者比例结构上，大部分为个人投资，机构投资比例低，从而导致非理性投资行为严重，投机现象存在，导致资金的时间价值为负值。
[Abstract]:As one of the core contents of the modern financial theory, the capital asset pricing model (CAPM) is first put forward by Sharp, Lintner, 1965) and Moxinman (1966) on the basis of Markowitz's mean-variance theory. Then Blackan Jensen and Scholes put forward the famous two-stage regression method of BJS and Roll criticism of Fama-Macbeth and the famous three-factor model proposed by Fama-French. Many scholars have been relaxing the harsh assumptions of standard CAPM model. The intertemporal capital asset pricing model (ICAPMN) is gradually developed and formed, which is based on the lower venture capital asset pricing model (D-CAPMN), the three-factor model (three-factor model) and the return Beta method capital asset pricing model (return 尾 CAPMM). In 1990s, the Shanghai Stock Exchange of China and the Shenzhen Stock Exchange were established one after another, and 20 years since their establishment, they have been applied to various fields of the bank. There are many institutional imperfections in the emerging markets compared with the foreign capital markets. Therefore, the applicability of CAPM model in China remains to be studied. This paper selects the monthly return rate of the sample stock of Shanghai 180 index from 2003 to 2012, divides the study interval into subinterval T1 from 2003-2007 and subinterval T2 from 2008 to 2008. The standard CAPM model, the lower risk CAPM model and the FF three-factor model were analyzed by using time series regression and two-stage regression under three factor grouping and 尾 value grouping. The CAPM model of return beta method and the three-factor model based on lower risk FF are comprehensively compared and analyzed, as well as the in-depth study on the relationship between risk and return of each combination. The following conclusions are drawn:. First, the method of measuring the system risk 尾 will affect the explanatory power of the model to the return rate and the overall fitting degree of the model. We know that the 尾 coefficient is an index to measure the volatility of a single asset relative to the market, and the calculation method of the 尾 value, There are many scholars at home and abroad, the common conclusion is that different methods will affect the overall effect of the model. Secondly, the influence of non-system risk on income is significant, which can be seen from the comparison of FF three-factor model and standard CAPM model, lower risk CAPM model and return Beta CAPM model. And the rate of return is inversely proportional to the scale factor and positively related to market value. Third, the adaptability of all models in Chinese stock market is not strong. In the comparison of multiple groups and multiple methods, the standard CAPM model and the return beta method CAPM model have similar explanatory power and are not very strong. Although the lower risk CAPM model and FF three-factor model and the three factor model based on the lower risk FF have been improved, the overall fitness and explanatory power are not very strong. 4th. In addition to the study on the return risk of the whole sample, most of the conclusions prove that the time value of the stock market in China is negative, which indicates that the capital market of our country is not mature enough, perfect, compared with the mature market of foreign countries. The current situation of the domestic stock market is that the information disclosure is not enough, the financial fraud is more, the heavy financing is less than the dividend, the information asymmetry, the noise transaction, the influence of the policy and so on, and the proportion structure of the investors. Most of them are personal investment, and the proportion of institutional investment is low, which leads to the serious irrational investment behavior, the existence of speculation, and the negative value of the time of capital.
【学位授予单位】：贵州财经大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：F830.42;F224

【参考文献】